Spring AI Alibaba 82｜Graph Core · LLM 流式输出 · 使用流式 ChatClient

Spring AI Alibaba 支持通过 ChatClient 进行流式输出。

import org.springframework.ai.chat.client.ChatClient;
import org.springframework.ai.chat.model.ChatResponse;
import reactor.core.publisher.Flux;

// 使用流式输出
Flux<ChatResponse> flux = chatClient.prompt()
      .user("tell me a joke")
      .stream()
      .chatResponse();

// 订阅流式响应
flux.subscribe(
      response -> {
          String content = response.getResult().getOutput().getText();
          System.out.print(content);
      },
      error -> System.err.println("Error: " + error.getMessage()),
      () -> System.out.println("
Stream completed")
);

使用 Reactor 的阻塞式处理

Flux<ChatResponse> flux = chatClient.prompt()
      .user("tell me a joke")
      .stream()
      .chatResponse();

// 使用 Reactor 的阻塞式处理
flux.collectList().block().forEach(response -> {
  System.out.println("Received: " + response.getResult().getOutput().getText());
});

输出示例:

Sure, here's a joke for you:

Why don't scientists trust atoms?

Because they make up everything!
Stream completed

创建带流式输出的 Graph 节点

参考 节点流式输出文档 获取完整示例。

import com.alibaba.cloud.ai.graph.OverAllState;
import com.alibaba.cloud.ai.graph.action.NodeAction;
import org.springframework.ai.chat.client.ChatClient;
import reactor.core.publisher.Flux;

import java.util.Map;

public class StreamingAgentNode implements NodeAction {

  private final ChatClient chatClient;

  public StreamingAgentNode(ChatClient.Builder builder) {
      this.chatClient = builder.build();
  }

  @Override
  public Map<String, Object> apply(OverAllState state) {
      String userMessage = (String) state.value("query").orElse("Hello");

      // 使用流式输出
      Flux<String> contentFlux = chatClient.prompt()
              .user(userMessage)
              .stream()
              .content();

      return Map.of("answer", contentFlux);
  }
}

配置和运行

import com.alibaba.cloud.ai.graph.StateGraph;
import com.alibaba.cloud.ai.graph.OverAllState;
import com.alibaba.cloud.ai.graph.CompiledGraph;
import org.springframework.ai.chat.client.ChatClient;

// 配置 Graph
StateGraph graph = new StateGraph(keyStrategyFactory)
  .addNode("agent", new StreamingAgentNode(chatClientBuilder))
  .addEdge(StateGraph.START, "agent")
  .addEdge("agent", StateGraph.END);

CompiledGraph compiledGraph = graph.compile();

// 执行
Map<String, Object> input = Map.of("query", "Hello");
OverAllState result = compiledGraph.invoke(input);

System.out.println("Final result: " + result.value("answer").orElse(""));

• 节点流式输出 – 完整的流式输出示例
• 快速入门 – Graph 基础使用
• Spring AI 文档 – Spring AI 官方文档

本文整理自 Spring AI Alibaba 官方文档《LLM 流式输出》：https://java2ai.com/docs/frameworks/graph-core/examples/llm-streaming-springai